Outer Step 0. Inner Step 137. Batch Size 296

config.json

@@ -268,23 +268,23 @@
     "AutoModelForCausalLM": "distributed/optimized-gpt2-500m--modeling_gpt_optimized.GPTOptim"
   },
   "block_list": [
-    5141195,
-    5141220,
-    5141249,
-    5141280,
-    5141295,
-    5141319,
-    5141337,
-    5141349,
-    5141367,
-    5141380
+    5141404,
+    5141409,
+    5141413,
+    5141429,
+    5141444,
+    5141462,
+    5141474,
+    5141484,
+    5141491,
+    5141500
   ],
   "block_size": 1024,
   "bos_token_id": 50256,
   "embd_pdrop": 0.1,
   "eos_token_id": 50256,
   "initializer_range": 0.02,
-  "inner_step": 130,
+  "inner_step": 137,
   "inner_steps": 0,
   "last_allreduce_block": 5072405,
   "layer_norm_epsilon": 1e-05,

inner_optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:20ccc722cb38218dfbb0a21bb65bec36e541eeb96a8c3c7db5982c10fffbc7ef
+oid sha256:665b82fa91a1ee64f3de476480b8dcbb5a086b53f4750bf2c9f9cd678663ec14
 size 8081781770

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:587fbc1044c2573f3df63f378c833b56df596e6f47c8f0d40c2bd20077f5fbf6
+oid sha256:c1de7776c95ba55f926fdf0c2f63e804fb4ff581b5cfc43f6a63b175c68e2048
 size 4040701744